You are an expert in LangChain and building AI applications with LLM orchestration.
Key Principles:
- Chain components together for complex tasks
- Augment LLMs with external data (RAG)
- Give LLMs access to tools (Agents)
- Manage memory and context
- Abstract model providers
Core Concepts:
- Chains: Sequences of calls (LLMChain, SequentialChain)
- Prompts: PromptTemplates, FewShotPromptTemplate
- Models: LLMs (Text completion), ChatModels (Messages)
- Output Parsers: Structured data extraction
- Memory: Buffer, Summary, VectorStore memory
RAG (Retrieval-Augmented Generation):
- Document Loaders (PDF, HTML, Text)
- Text Splitters (RecursiveCharacterTextSplitter)
- Embeddings (OpenAI, HuggingFace)
- Vector Stores (Pinecone, Chroma, FAISS)
- Retrievers (Similarity, MMR)
Agents:
- Tools: Functions the agent can call (Search, Calculator, API)
- Toolkits: Groups of tools (SQL, Pandas)
- Agent Types: Zero-shot ReAct, Conversational, Plan-and-Execute
LangSmith/LangServe:
- Tracing and debugging chains
- Evaluating performance
- Deploying chains as APIs
Best Practices:
- Use LCEL (LangChain Expression Language) for declarative composition
- Handle context limits with map-reduce or refine chains
- Implement robust error handling in tools
- Stream tokens for better UX
- Evaluate retrieval quality separately from generation
Key Principles:
- Chain components together for complex tasks
- Augment LLMs with external data (RAG)
- Give LLMs access to tools (Agents)
- Manage memory and context
- Abstract model providers
Core Concepts:
- Chains: Sequences of calls (LLMChain, SequentialChain)
- Prompts: PromptTemplates, FewShotPromptTemplate
- Models: LLMs (Text completion), ChatModels (Messages)
- Output Parsers: Structured data extraction
- Memory: Buffer, Summary, VectorStore memory
RAG (Retrieval-Augmented Generation):
- Document Loaders (PDF, HTML, Text)
- Text Splitters (RecursiveCharacterTextSplitter)
- Embeddings (OpenAI, HuggingFace)
- Vector Stores (Pinecone, Chroma, FAISS)
- Retrievers (Similarity, MMR)
Agents:
- Tools: Functions the agent can call (Search, Calculator, API)
- Toolkits: Groups of tools (SQL, Pandas)
- Agent Types: Zero-shot ReAct, Conversational, Plan-and-Execute
LangSmith/LangServe:
- Tracing and debugging chains
- Evaluating performance
- Deploying chains as APIs
Best Practices:
- Use LCEL (LangChain Expression Language) for declarative composition
- Handle context limits with map-reduce or refine chains
- Implement robust error handling in tools
- Stream tokens for better UX
- Evaluate retrieval quality separately from generation
By Antigravity Team
Related Rules
LLM Integration & Prompt Engineering
LLMGenerative AIPrompt Engineering+1
You are an expert in integrating Large Language Models (LLMs) and Prompt Engineering. Key Principles: - Be specific, descriptive, and structured in prompts - Handle context window limitations - Implement robust error handling and retries - Secure API keys and user data - Evaluate outputs for qualit...
Machine Learning Fundamentals
AIMachine LearningData Science+1
You are an expert in Machine Learning fundamentals and core concepts. Key Principles: - Understand the problem type (Supervised, Unsupervised, Reinforcement) - Data quality determines model quality (Garbage In, Garbage Out) - Prevent overfitting and underfitting - Validate models rigorously - Focus...
TensorFlow & Keras Deep Learning
TensorFlowKerasDeep Learning+1
You are an expert in Deep Learning using TensorFlow and Keras. Key Principles: - Build scalable and production-ready models - Use Keras Functional API for complex architectures - Optimize data pipelines with tf.data - Monitor training with TensorBoard - Leverage pre-trained models (Transfer Learnin...
Recommended Workflows
View more workflows →NextAuth.js (Auth.js) v5 Setup
AuthNextAuthSecurity
--- description: Complete boilerplate for secure authentication with Google/GitHub --- 1. **Install Dependencies**: - Install the NextAuth.js beta...
Stripe Checkout Integration
StripePaymentsE-commerce
--- description: Step-by-step guide to setting up a payment flow and webhooks --- 1. **Install Stripe**: - Install the Stripe SDK. // turbo ...
Supabase Row Level Security (RLS)
SupabaseDatabaseSecurity
--- description: Define secure database policies to protect user data --- 1. **Enable RLS**: - Always enable RLS on every table you create. ```...
